Dual-band miniaturized composite right left handed transmission line ZOR antenna for microwave communication with machine learning approach

Abstract

This article presents a dual-band miniaturized Composite Right-Left-Handed Transmission Line (CRLH-TL) in an open-ended terminal, employing the Machine Learning (ML) technique. The CRLH-TL antenna is designed on the FR4 epoxy substrate. The substrate size is 0.31λ0 × 0.09 λ0, where λ0 is the free space wavelength. The proposed antenna offers dual-band functionality with resonant frequencies at 2.49 GHz and 5.33 GHz. The measured dual-band impedance bandwidths are 14.46 % and 14.74 %, with gains of 0.85 dB and 1.67 dB, and radiation efficiencies of 91.78 % and 95.45 % obtained at resonating frequencies of 2.49 GHz and 5.33 GHz, respectively. The proposed antenna also offers bipolar-type radiation patterns in the E-plane, and omnidirectional radiation patterns in the H-plane, along with compactness and constant gain. Several ML methods, including Random Forest (RF), Decision Tree (DT), K-Nearest Neighbour (KNN), Extreme Gradient Boosting (XGB), and Artificial Neural Network (ANN), are used to optimize the antenna. Compared to other ML algorithms, RF ML techniques estimate reflection coefficient S11 with an accuracy of above 98 %. The proposed antenna is utilized in WLAN (5.15–5.35, 5.47–5.725 GHz) and Wi-MAX (5.2–5.8 GHz) microwave applications.